Allogeneic hematopoietic stem cell transplantation (alloSCT) can be a curative therapy for hematologic malignancies and inherited and acquired disorders of blood cells. Alloreactive graft T cells mediate the graft-vs- leukemia (GVL) effect and contribute to immune reconstitution. However, they also attack normal host tissues causing graft-vs-host disease (GVHD). A central goal of alloSCT research has been to discover methods of minimizing and treating established GVHD with relative sparing of GVL and anti-pathogen immunity. This has been a challenge as most GVHD therapies delete or target basic T cell functions, though there are a few promising new relatively GVHD-specific approaches. New and more specific approaches, however, require a better understanding of GVHD. We considered the question of how GVHD is maintained despite persistent and unlimited antigen; whereas in other models, chronic T cell-antigen exposure results in exhaustion, anergy and deletion. One possibility is that GVHD is sustained by the continuous generation of alloreactive effectors derived from secondary lymphoid tissues (SLT), which traffic to GVHD target tissues. An alternative possibility is, that once tissues are seeded with alloreactive effectors, GVHD is maintained locally without significant input from SLT-derived T cells. Whether GVHD is maintained locally and/or in SLT, another key question is whether there are specific subsets of T cells that replenish effectors despite constant antigen exposure; if identified, such cells would be ideal to target. We looked to antipathogen immunity for guidance on these questions. Robust antipathogen T cell responses occur in peripheral tissues, including skin, lung, bowel and vaginal mucosa, all GVHD targets. Such responses can lead to the generation within tissues of a newly described memory T cell (TM) subset called tissue resident memory cells (TRM) which do not circulate to other locations. Upon antigen rechallenge in tissues, TRM are rapidly activated, proliferate and differentiate into effectors. To experimentally test the hypothesis that GVHD is ?local?, we tracked the clonal progeny of single GVHD- inducing TCR transgenic (Tg) CD4+ T cells (TS1) and found them to be unequally distributed among GVHD target tissues and not in equilibrium with TS1 in SLT. These clones were capable of developing progeny with diverse phenotypes, suggesting multiple differentiation pathways being available after priming. Importantly, we found TS1 and alloreactive polyclonal T cells within GVHD target tissues with immunophenotypes and gene expression profiles in common with those reported for TRM-like cells. This proposal will test the fundamental hypotheses that GVHD is locally sustained and that maintenance within GVHD target tissues is fueled by TRM-like cells. We will do so in models wherein CD4 and CD8 cells mediate GVHD. We will apply rigorous statistical and mathematical approaches to test this idea. If the Aims confirm our hypotheses, the long-term goal is to develop methods to specifically delete or impair these TRM-like cells, first in mouse models, and then in the clinic.